DeepSeek V4 Flash 0731 vs GLM 5.2: price, benchmarks and speed

Side by side

	DeepSeek V4 Flash 0731	GLM 5.2
Lab	DeepSeek	Z.AI
Listed	Jul 31, 2026	Jun 16, 2026
Input per 1M tokens	$0.088	$0.009
Output per 1M tokens	$0.45	$10.20
Cached input per 1M	$0.052	$0.0075
Blended (3:1) per 1M	$0.18	$2.56
Context window	1M	1M
Max output	944K	944K
Knowledge cutoff	-	-
Intelligence index	34.5	34
Coding index	69.1	68.8
Reasoning	Yes	Yes
Reads images	No	No
Tool use	Yes	Yes
Structured output	Yes	Yes

Every benchmark both are scored on

Each model's best score per benchmark, with its source. Scores from different sources or settings are not always comparable; each links to where it was published.

Benchmark	DeepSeek V4 Flash 0731	GLM 5.2	Leader
AIME	94.4%Epoch AI	28.9%Epoch AI	DeepSeek V4 Flash 0731
ARC-AGI-1	89%Epoch AI	77%Epoch AI	DeepSeek V4 Flash 0731
ARC-AGI-2	61.4%Epoch AI	22.8%Epoch AI	DeepSeek V4 Flash 0731
Artificial Analysis Agentic Index	41Artificial Analysis and OpenRouter	38.4Artificial Analysis and OpenRouter	DeepSeek V4 Flash 0731
Artificial Analysis Coding Index	69.1Artificial Analysis and OpenRouter	68.8Artificial Analysis and OpenRouter	DeepSeek V4 Flash 0731
Artificial Analysis Intelligence Index	34.5Artificial Analysis and OpenRouter	34Artificial Analysis and OpenRouter	DeepSeek V4 Flash 0731
Chess Puzzles	32%Epoch AI	22%Epoch AI	DeepSeek V4 Flash 0731
Design Arena: 3D	1205 EloDesign Arena	1284 EloDesign Arena	GLM 5.2
Design Arena: ASCII Art	1088 EloDesign Arena	1188 EloDesign Arena	GLM 5.2
Design Arena: Code	1228 EloDesign Arena	1285 EloDesign Arena	GLM 5.2
Design Arena: Data Viz	1180 EloDesign Arena	1286 EloDesign Arena	GLM 5.2
Design Arena: Game Dev	1207 EloDesign Arena	1266 EloDesign Arena	GLM 5.2
Design Arena: SVG	1180 EloDesign Arena	1207 EloDesign Arena	GLM 5.2
Design Arena: UI Component	1230 EloDesign Arena	1290 EloDesign Arena	GLM 5.2
Design Arena: Website	1236 EloDesign Arena	1283 EloDesign Arena	GLM 5.2
DTBench	90.9%Epoch AI	93.6%Epoch AI	GLM 5.2
FrontierCode	18.8Epoch AI	24.5Epoch AI	GLM 5.2
FrontierMath Tier 4 (v2)	24.4%Epoch AI	29.3%Epoch AI	GLM 5.2
FrontierMath Tiers 1-3 (v2)	57.5%Epoch AI	42.5%Epoch AI	DeepSeek V4 Flash 0731
GPQA Diamond	86.8%Artificial Analysis and OpenRouter	84.7%Artificial Analysis and OpenRouter	DeepSeek V4 Flash 0731
LMArena (Coding)	1456 EloLMArena	1487 EloLMArena	GLM 5.2
LMArena (Overall)	1432 EloLMArena	1470 EloLMArena	GLM 5.2
LMCA	41.7Epoch AI	45.8Epoch AI	GLM 5.2
Mystery Game Puzzles	34%Epoch AI	15%Epoch AI	DeepSeek V4 Flash 0731
ProofBench	56%Epoch AI	35%Epoch AI	DeepSeek V4 Flash 0731
SciCode	49.9%Epoch AI	51.2%Epoch AI	GLM 5.2
SimpleBench	61.1%Epoch AI	58.8%Epoch AI	DeepSeek V4 Flash 0731
SimpleQA Verified	33.6%Epoch AI	34.2%Epoch AI	GLM 5.2
tau-bench (Airline)	74%Artificial Analysis and OpenRouter	71.3%Artificial Analysis and OpenRouter	DeepSeek V4 Flash 0731
WebDev Arena	1580 EloEpoch AI	1593 EloEpoch AI	GLM 5.2
WeirdML	63%Epoch AI	70.1%Epoch AI	GLM 5.2

More comparisons

Where these figures come from

Prices are GenMagic's own, what the studio and the API charge, recomputed from each provider's price list every few minutes. The indices are the Artificial Analysis indices, and every benchmark score links to its source, re-read daily. Measured speed is GenMagic's own generation records, in aggregate. See DeepSeek models, Z.AI models, every LLM's price and our sources.